Although Option B+ is being rolled out in many developing countries for the prevention of mother-to-child HIV transmission (PMTCT), robust evidence of its population-level effectiveness and cost-effectiveness is currently lacking. This is complicated by the fact that the impact of Option B+ may vary according to whether PMTCT and antiretroviral therapy (ART) services are already successfully integrated in health facilities. A population- level effectiveness study would determine the HIV-free survival rate and/or the mother-to-child transmission of HIV (MTCT) rate among the entire community of infants and not just among those whose mothers accessed PMTCT services. Another outstanding issue is the extent to which it will be possible to retain women on lifelong ART, especially those who start ART during pregnancy and who are asymptomatic. Our proposed study will fill these research gaps using representative community-level serosurvey data from Zimbabwe. In 2012 our team conducted a community survey of over 9,000 mother-infant pairs that were born before the implementation of Option A in Zimbabwe; preliminary results revealed that HIV prevalence among mothers was 12.5% and MTCT was 9% among infants aged 9-18 months. Between October 2013 and September 2014, Zimbabwe will roll out Option B+, requiring integration of PMTCT and ART services in all 1,560 facilities. In 2014, we will conduct a post-Option A / pre-Option B+ survey of 10,000 mother-infant pairs using methodology similar to that of our 2012 survey. In 2017 we will survey 13,000 mother-infant pairs and will assess the impact of Option B+ using two strategies. First, we will compare outcomes from 2017 versus 2012, permitting a comparison of Option B+ to the standard of care prior to the implementation of Option A (Aim 1a). Second, we will compare outcomes from 2017 to those from 2014, permitting a direct assessment of the value- added of Option B+ over Option A (Aim 1b). Notably, HIV-free survival among HIV-exposed infants takes into account both HIV-infection and mortality, which is critical given the high mortality rates experienced by HIV- exposed and HIV-infected infants. By focusing on infants born 9-18 months prior to the surveys we capture transmissions occurring during pregnancy, labor and breastfeeding. In addition, we will examine impact heterogeneity by the level of integration of PMTCT and ART services at the facility and other facility characteristics (Aim 2). The 2017 community-based survey will also include 1,800 mothers of 19-36 months old infants, to assess HIV-infected mothers' retention in ART services after weaning (Aim 3). In addition to household-level data, we will collect comprehensive facility-level data to assess cost- effectiveness of Option B+ (Aim 4). Our results will have direct relevance for Zimbabwe and other developing countries, as they will inform evidence-based decisions for Ministries of Health that are considering Option B+, irrespective of whether they implemented previous WHO guidelines including Option A.